Sign up Calendar Latest Topics Donate
 
 
 


Reply
  Author   Comment  
RickP330

Member
Registered:
Posts: 30
Reply with quote  #1 

Hi Guys,

I am having difficulty understanding how this data collector is collecting data.  I want to set my own limits but it appears to be taking several samples per collection and auto ranging the data.

 

In my “Data Collection set up” I appear to be able to set my Lines of Resolution (LOR) by clicking the drop down menu and selecting 1600 from the field “Resolution”. 

 

To set my Fmax It appears I have to enter the “vibration location” and edit the “vibration readings” tab.  The default set up is “Low Range” = 300 Hz  “High Range” – 8000 Hz and Demod = 600 Hz.  My Time waveform shows 5 seconds of data. 

 

My Low range FFT’s goes out to 12X and High range goes to 300X.  In this example my RPM is set at 1705. 

 

If T= LOR/Fmax then it appears to be using the “Low Range” setting of 300 Hz and giving me 5 seconds of data.  If I only want to look at a handful of revolutions on the Time waveform I have to zoom into .2 seconds or so.  The data is too course this low.

 

I have to trick the collector and set the “Low Range” to 8000 Hz to get a pretty .2 Second time waveform, but then my FFT plots are off the chart (300X on the low range and 600x (17 kHz) on the high range.

 

So What I have to do is duplicate every machine and collect data as normal (to view the FFT and a useless TW) and then do it again with a “trick” set of parameters to get a nice Time Waveform (and useless FFT) picture.  I am doubling the data, and collection points to try and get what I want. 

 

Am I doing something wrong?  Is there some way I can set this thing to give me the Time Waveform I desire?

 

I’m sorry for the long winded post to ask such a direct question, but writing it out really helped me understand the problem here.  Much Thanks in advance,

Rick

 

Edwin

Avatar / Picture

Sr. Member
Registered:
Posts: 131
Reply with quote  #2 
Indeed the length of the timewave is defined by the lines and freq settings: number of lines devided by low pass frequency (Hz) is the timewave length in seconds.
To measdure a longer time I also use an extra acceleration measurement, indeed with a useless fft. (like 3200 lines and 300 Hz = 10.7 seconds)

__________________
= Cost Effective Asset Management =
RickP330

Member
Registered:
Posts: 30
Reply with quote  #3 
Okay,
So this is business as usual I guess?
Seems like a waste of data and time....
RP
JuddJones

Sr. Member
Registered:
Posts: 384
Reply with quote  #4 
Change is coming. In the future no one will average, and everybody will take 1 time waveform and post process all the other stuff they need.
RickP330

Member
Registered:
Posts: 30
Reply with quote  #5 
Oh, I just found in the set up, it processes the TW on both the High and Low Resolution settings.  I can turn off the low it's useless to me at 5 seconds long.  So I had the data I needed and no need to take an extra set of readings.  I only realized this on the route this morning when I was actually watching what it was reading / processing...
RP
MarkL

Avatar / Picture

Sr. Member
Registered:
Posts: 1,121
Reply with quote  #6 
Quote:
Originally Posted by Big J
Change is coming. In the future no one will average, and everybody will take 1 time waveform and post process all the other stuff they need.


Ive started to use that recently with the recorder function on microlog and post process in ARM, not exactly the same, but same principle, I often wish we just had to take one decent twf and post process the crap out of it after. ;-)
That is what these guys seem to be at as well.

Previous Topic | Next Topic
Print
Reply

Quick Navigation:

Easily create a Forum Website with Website Toolbox.